Clusters of galaxies are potentially important targets for indirect searches for dark matter annihilation. Here we reassess the detection prospects for annihilation in massive halos, based on a statistical investigation of 1743 clusters in the new Meta-Catalog of X-ray Clusters. We derive a new limit for the extra-galactic dark matter annihilation background of at least 20% of that originating from the Galaxy for an integration angle of 0.1 deg. The number of clusters scales as a power law with their brightness, suggesting that stacking may provide a significant improvement over a single target analysis. The mean angle containing 80% of the dark-matter signal for the sample is ~0.15 deg, indicating that instruments with this angular resolution or better would be optimal for a cluster annihilation search based on stacking. A detailed study based on the Fermi-LAT performance and position-dependent background, suggests that stacking may result in a factor ~2 improvement in sensitivity, depending on the source selection criteria. Based on the expected performance of CTA, we find no improvement with stacking, due to the requirement for pointed observations. We note that several potentially important targets: Opiuchius, A2199, A3627 (Norma) and CIZAJ1324.7-5736 may be disfavoured due to a poor contrast with respect to the Galactic dark-matter signal. The use of the homogenised MCXC meta-catalogue provides a robust ranking of the targets, although the absolute value of their signal depends on the exact dark matter substructure content. For conservative assumptions, we find that galaxy clusters (with or without stacking) can probe down to 1e-25-1e-24 cm3/s for dark matter masses in the range 10 GeV-100 GeV. For more favourable substructure configurations, ~1e-26 cm3/s may be reached.